Acoustics and EVAC I

Correct installation and measurement of the systems for sound evacuation in the case of emergencies.


The emergency audio evacuation systems (EVAC) have been discussed greatly in recent years amongst employees. The main reason is that it’s about security systems and, therefore, they are subject to very specific supply rules, installation and set-up procedures (which is rather unusual in the audio sector), but also because they now actually cover a market segment which has become a real business for the whole sector (for manufacturers, to installers, system integrators and technicians). The philosophy of these systems is based on the certainty of being able to hear the evacuation signals in case of danger. To do this, the system has to comply to specific rules and possess certifications of various kinds for speakers, wiring and control units, which have been widely discussed over the years. As important as all these aspects are, they do not guarantee that the goal described above will be achieved (understanding the message), because they do not take in account the properties of the environment where the evacuation signals will have to be transmitted. In fact, the existing regulations state that, for each installation, at every point, the acoustic requirements are upheld and, in particular, the intelligibility of speech. This framework - STI (Speech Transmission Index and its “surrogates”, such as the STIPA etc.) -changes depending on the signal/noise ratio, so, for the same sound source output, its value diminishes if the reverberation of the space increases, or if there is a high level of background noise. Experience, however, teaches us that using different audio systems can partially improve intelligibility. Direct speakers can, for example, improve the STI, but there are physical limits meaning that in particularly problematic environments, the reverberant field is so dense that it is virtually impossible to achieve even minimal results; in these cases, the only thing to do is to improve the environment from an acoustic point of view. In some drastic situations, “aggressive treatment” leads to the insertion of more sound sources, resulting in a particularly problematic situation: in a reverberant environment, the overlapping of multiple direct emissions with reflex/ reverberated emissions, involves a phase shift of the original signal, which affects the intelligibility of speech. At the side, you will see a table of STI values, great for various situations.


STI- Speech Transmission Index


The STI method comes from the study of the smallest fragments of speech signals, called phonemes, each one characterised by its spectrum of a specific frequency. Transparency requires that the spectral differences of the phonemes are maintained: they are identified by the envelope function between various frequency bands, which describes the temporal fluctuations of the speech signal strength for a certain frequency band. In fact, there is one, and one only, form of the envelope function for each specific sequence of phonemes. Any distortion of the envelope of speech, due to, for example, noise or reverberation, is shown as a reduction of spectral differences between the phonemes; in other words, it is imitated in a reduced level of fluctuations of the envelope function. From the comparison between the spectra envelope, obtained directly from the speaker, and the corresponding spectra, obtained via a transmission path, you get the reduction of fluctuations, due to the transmissions channel (in our case the environment). This reduction leads to defining the transfer function of the modulation (Modulation Transfer Function -MTF) that represents the reduction of the modulation index as the function of the modulation frequency.

Screen Shot 2017-04-08 at 14.56.42.png


Regulations and Measurements

The current regulations (UNI ISO 7240 CEI EN 60849) require the minimum values of STI (0.50 as the average value and 0.45 as the minimum) to be obtained in various measurement positions, whose minimum figure is relative to the size of the area where the EVAC system is operating. Furthermore, it requires an annual inspection of the system and new implementation of all the acoustic measurements. There is, however, some confusion surrounding the role of the technician and the way in which measurements should be made. The regulation requires a “certified technician” and the only professional figure in Italy who can perform certified acoustic measurements is the Competent Technician in Environmental Acoustics. The measurements, on the other hand, can be of two types, direct or indirect, and for more details on the STI parameter, please see the appropriate box. The direct method for measuring the STI is shown in paragraph 5 of the regulation IEC 60268-16:2011 and consists of, in the stimulus of the environment, 98 separate test signals which use 14 different modulation frequencies, for seven octaves of frequency. Every test signal contains a single modulation frequency for a single frequency band, while the other bands do not contain signals. With an average of 10 seconds for each signal, a complete measurement of the FULL STI takes around 15 minutes (!). The STIPA index (STI for Public Address which, as mentioned above, is not “better” than the STI, but is a simplification of it) instead consists of a single test signal, with a predefined set of two modulations in each of the seven bands of octaves. The 14 modulation frequencies are generated simultaneously, thus the measurements only takes 10-15 seconds. However, the modulation transfer function (MTF) can also be calculated with Schroeder’s method, once it has received the impulse response from a computer: this is the indirect method. Although there are limitations relating to the applicability of this method, there are also many studies that claim it is better compared to the direct method. Furthermore, the measurement time is very quick.


Measuring the STI

Heated discussions surround the topic of what tools are needed for measuring the STI. Almost all commercial sound level meters for sound detection in class 1, or rather, for measurements from laboratories on on-site , in defined acoustic conditions, they are not suitable or adaptable for the measurements set out on the STI index. There are two main reasons for this: many sound level meters do not have a required signal generator and do not have an output, meaning that the synchronisation between the generated signal and the acquired signal is unworkable (essential conditions for the impulse response). Furthermore, internally they often have a software which can calculate the transfer matrix and also include the background noise in the measurements, measured at another time. Therefore, the state of the art is such that professionals do not have the commercial tools particularly suitable for the purpose, but the measurement systems used come from the audio professional world, rather than from environmental acoustic measurement field. It is, for example, easier to transform a system for measurement and control of the transducers or speakers in an STI measurement system, rather than doing so with a traditional sound level meter. There are various softwares (apps, for example, for iPhones or tablets) which allow such measurements, but according to Italian laws, it is not easy to insert this measurement chain into class 1 and, therefore, make it stand in accordance to the law. Among the “certified” systems, we bring to you the NTI XLII for the direct measurement and the CLIO from Audiomatica for indirect measurement (and apparently, in the near future, also for direct measurement).



  • Laura Meucci: Sistemi di sicurezza EVAC di Segnalazione vocale per l’evacuazione di grandi ambienti: progettazione, messa in funzione e collaudo (con il metodo STI), in ambito acustico, relativi ad alcuni studi e progetti in corso d’opera; Degree thesis in Engineering.
  • European legislation EN 60849, from April 1998, Sound systems for emergency purposes, recognised in Italy in 1999 as the regulation CEI EN 60849 Sistemi elettroacustici applicati ai servizi di emergenza.
  • Legislation UNI ISO 7240-19 Fire detection and alarm systems - Part 19: Design, installation, commissioning and service of sound systems for emergency purposes, Sistemi fissi di rivelazione e di segnalazione allarme d'incendio - Parte 19: Progettazione, installazione, messa in servizio, manutenzione ed esercizio dei sistemi di allarme vocale per scopi d'emergenza.
  • European legislation EN 54 Fire detection and fire alarm systems, in particular parts 2, 4, 16, 24, which the 7240-19 makes specific reference to equipment making up 17 the sound system.